Files

292 lines
9.9 KiB
Python
Raw Permalink Normal View History

2024-04-02 14:11:37 +05:30
import json
2024-06-13 13:33:27 -05:00
import security
2024-03-27 18:59:29 +05:30
import streamlit as st
from streamlit_extras.add_vertical_space import add_vertical_space
import os
import streamlit as st
import pandas as pd
from io import StringIO
from datetime import datetime
import boto3
import util
import requests
2024-03-28 17:59:41 +05:30
from sf_conn import get_secret, save_to_sf
2024-04-02 14:11:37 +05:30
from io import StringIO, BytesIO
2024-06-13 14:34:06 -05:00
import time
from sf_conn import get_client_names, insert_upload_logs
2024-05-24 07:01:29 -05:00
from constants import USER_LIST
from util import logger
2024-03-27 18:59:29 +05:30
(REDIRECT_URI, create_batch_url, doczy_pipeline) = util.load_page_details(0)
2024-03-27 18:59:29 +05:30
2024-05-23 10:20:49 -05:00
user_list = USER_LIST
2024-09-30 12:21:29 +01:00
if "uploading" not in st.session_state:
st.session_state.uploading = False
2024-09-30 12:21:29 +01:00
if "upload_key" not in st.session_state:
st.session_state.upload_key = 0
if "file_list" not in st.session_state:
2024-06-13 14:34:06 -05:00
st.session_state.file_list = []
2024-09-30 12:21:29 +01:00
if "show_batchID" not in st.session_state:
2024-06-13 14:34:06 -05:00
st.session_state.show_batchID = False
2024-09-30 12:21:29 +01:00
if "landing_zone" not in st.session_state:
st.session_state.landing_zone = "contracts-landing-zone"
if "batch_id" not in st.session_state:
st.session_state.batch_id = "failed_cases"
if "client_bucket" not in st.session_state:
st.session_state.client_bucket = "default_bucket"
def insert_upload_logs(batch_id, client_name, file_name, upload_datetime, upload_user):
"""
Input: batch_id, client_name, file_name, upload_datetime, upload_user
Output: status of the insert query
"""
try:
2024-09-30 12:21:29 +01:00
return "Log inserted successfully"
except Exception as e:
return e
2024-03-27 18:59:29 +05:30
2024-09-30 12:21:29 +01:00
st.set_page_config(layout="wide")
2024-06-13 14:34:06 -05:00
# # # Sidebar contents
with st.sidebar:
st.title("Doczy.AI ™")
st.markdown(
"""
## About
This app extracts data from contracts
2024-03-27 18:59:29 +05:30
2024-06-13 14:34:06 -05:00
"""
)
add_vertical_space(15)
# st.write("Doczy")
2024-03-27 18:59:29 +05:30
# AARETE LOGO
2024-09-30 12:21:29 +01:00
x, y, z = st.columns([15, 2, 15])
2024-06-14 09:52:01 -05:00
with y:
2024-09-30 12:21:29 +01:00
st.image("aaretelogo.png")
2024-06-13 14:34:06 -05:00
2024-09-30 12:21:29 +01:00
hide_img_fs = """
2024-06-14 09:52:01 -05:00
<style>
button[title="View fullscreen"]{
visibility: hidden;}
</style>
2024-09-30 12:21:29 +01:00
"""
2024-06-14 09:52:01 -05:00
st.markdown(hide_img_fs, unsafe_allow_html=True)
2024-03-27 18:59:29 +05:30
2024-09-30 12:21:29 +01:00
_, c1 = st.columns([5, 1])
2024-03-27 18:59:29 +05:30
try:
2024-04-02 14:11:37 +05:30
util.setup_page(REDIRECT_URI)
2024-04-15 17:02:59 -05:00
except Exception as e:
st.write(f"SSO Failed = {e}")
2024-09-30 12:21:29 +01:00
st.session_state["user_info"] = {
"mail": "maamseek@aarete.com",
"displayName": "Mayank Aamseek",
} # RECOMMENDATION - Remove Line after development.
2024-09-27 13:13:04 +01:00
try:
2024-04-05 12:15:54 +05:30
c1.write(f"User: **{st.session_state.user_info['displayName']}**")
2024-09-30 12:21:29 +01:00
user_mail = st.session_state.user_info["mail"]
except (
KeyError
) as e: # RECOMMENDATION - Not Working as Expected. Unreachable Code because the above Exception handles everything
2024-04-05 12:15:54 +05:30
st.write("Session Expired.")
auth_url = security.get_auth_url(REDIRECT_URI)
2024-09-30 12:21:29 +01:00
st.markdown(
f"<a href='{auth_url}' target='_self'>Sign In</a>", unsafe_allow_html=True
)
2024-04-09 14:06:14 +05:30
st.stop()
2024-03-27 18:59:29 +05:30
2024-09-30 12:21:29 +01:00
s3_client = boto3.client(
"s3", # RECOMMENDATION - We can make use of AWS Access Key for enabling this in local testing. Will help a lot.
2024-05-28 17:32:36 -05:00
region_name="us-east-2",
)
2024-09-30 12:21:29 +01:00
client_list = [
"doczy-ai-client-1",
"Delaware First Health, Inc.",
"Community Health Choice, Inc",
"CareSource Network Partners LLC",
"HealthNet of Cali",
"Oklahoma Complete Health, Inc",
"HealthFirst",
"Molina Healthcare of TX",
"AvMed",
"Arizona Care1st",
"WellCare New Jersey",
]
2024-05-28 17:32:36 -05:00
2024-09-30 12:21:29 +01:00
# This is the list of client fetched from Snowflake
2024-05-28 17:32:36 -05:00
# TODO: Need to update the streamlit code to use the client names from this list
# And use the s3 paths to save the objects for the respective client
client_list, s3_paths = get_client_names()
logger.info(f"client_list={client_list}")
logger.info(f"s3_paths={s3_paths}")
2024-05-28 17:32:36 -05:00
client_s3_paths = dict(zip(client_list, s3_paths))
client_row = st.columns([0.1, 0.8])
with client_row[0]:
st.write("**Client Name**")
with client_row[1]:
2024-09-30 12:21:29 +01:00
client = st.selectbox(
"Client Name", (client_list), label_visibility="collapsed", index=None
)
2024-05-28 17:32:36 -05:00
client_bucket = client_s3_paths.get(client)
# to be deleted when client buckets are created
2024-06-20 16:45:31 -05:00
# client_bucket = 'doczyai-use2-d-cn1-s3-textract-processing-001'
2024-05-28 17:32:36 -05:00
file_row = st.columns([0.1, 0.8])
with file_row[0]:
st.write("**Upload Files**")
with file_row[1]:
2024-09-30 12:21:29 +01:00
file_list = st.file_uploader(
"Upload",
type=["docx", "tiff", "pdf"],
accept_multiple_files=True,
label_visibility="collapsed",
help="Only PDF, TIFF and DOCX file formats are supported.",
disabled=st.session_state.uploading,
key=st.session_state.upload_key,
)
2024-05-28 17:32:36 -05:00
add_vertical_space(2)
2024-09-30 12:21:29 +01:00
df = pd.DataFrame(columns=["Contract Name"])
df["Contract Name"] = file_list
2024-05-28 17:32:36 -05:00
file_names = []
buttons = st.columns([0.4, 0.4, 0.2])
2024-09-30 12:21:29 +01:00
def set_uploading_state():
if not client == None and not len(file_list) == 0:
2024-06-13 14:34:06 -05:00
st.session_state.file_list = file_list
st.session_state.upload_key += 1
st.session_state.uploading = True
2024-09-30 12:21:29 +01:00
2024-05-28 17:32:36 -05:00
with buttons[1]:
2024-06-13 14:34:06 -05:00
if st.button("Create Batch", on_click=set_uploading_state):
file_list = st.session_state.file_list
if client == None:
st.error("No Client Name Selected.")
elif len(file_list) == 0:
st.error("No Files Selected.")
2024-05-28 17:32:36 -05:00
else:
2024-09-30 12:21:29 +01:00
with st.spinner("Running..."):
myobj = {"client-bucket-name": client_bucket}
response = requests.post(create_batch_url, json=myobj)
2024-06-13 14:34:06 -05:00
if response.status_code >= 200 and response.status_code < 300:
try:
2024-09-30 12:21:29 +01:00
batch_id = json.loads(json.loads(response.text)["body"])[
"batch_id"
]
landing_zone = json.loads(json.loads(response.text)["body"])[
"landing_zone"
]
landing_zone = "contracts-landing-zone/"
2024-06-13 14:34:06 -05:00
except:
# st.write(myobj)
# st.write(response.text)
st.write("Internal Error. Reach out to Doczy.AI Team.")
2024-09-30 12:21:29 +01:00
batch_id = "failed_cases"
landing_zone = "contracts-landing-zone/"
2024-06-13 14:34:06 -05:00
else:
st.error("Failed")
for uploaded_file in file_list:
stringio = BytesIO(uploaded_file.getvalue())
stringio.seek(0)
2024-09-30 12:21:29 +01:00
s3_client.put_object(
Bucket=client_bucket,
Body=stringio.getvalue(),
Key=landing_zone + batch_id + "/" + str(uploaded_file.name),
)
s3_client.put_object_tagging(
Bucket=client_bucket,
Key=landing_zone + batch_id + "/" + str(uploaded_file.name),
Tagging={"TagSet": [{"Key": "BatchId", "Value": batch_id}]},
)
2024-09-30 12:21:29 +01:00
# TODO: Test this insert function with snowflake
upload_log = insert_upload_logs(
batch_id,
client,
str(uploaded_file.name),
datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
user_mail,
)
2024-06-13 14:34:06 -05:00
st.write(upload_log)
2024-09-30 12:21:29 +01:00
2024-06-13 14:34:06 -05:00
file_names.append(str(uploaded_file.name))
st.session_state.uploading = False
st.session_state.show_batchID = True
st.session_state.client_bucket = client_bucket
st.session_state.batch_id = batch_id
st.session_state.landing_zone = landing_zone
2024-09-30 12:21:29 +01:00
st.rerun()
2024-06-13 14:34:06 -05:00
if st.session_state.show_batchID:
2024-06-17 15:12:02 -05:00
batch_id = st.session_state.batch_id
client_bucket = st.session_state.client_bucket
landing_zone = st.session_state.landing_zone
2024-06-13 14:34:06 -05:00
st.write("Batch ID Created - ")
2024-06-17 15:12:02 -05:00
st.code(f"{batch_id}")
2024-05-28 17:32:36 -05:00
st.write(f"Files uploaded to s3://{client_bucket}/{landing_zone}{batch_id}")
2024-06-14 09:40:45 -05:00
st.session_state.show_batchID = False
st.session_state.file_list = []
2024-09-30 12:21:29 +01:00
st.session_state.batch_id = "failed_cases"
st.session_state.client_bucket = "default_bucket"
st.session_state.landing_zone = "contracts-landing-zone"
2024-03-28 17:53:00 +05:30
# RECOMMENDATION - Rethink about how to best manage cache
2024-09-30 12:21:29 +01:00
# @st.cache_data
2024-03-28 17:53:00 +05:30
# def convert_df(df):
# return df.to_csv(index=False).encode('utf-8')
2024-03-27 18:59:29 +05:30
2024-03-28 17:53:00 +05:30
# csv = convert_df(df)
# df = df.reset_index() # make sure indexes pair with number of rows
2024-03-27 18:59:29 +05:30
# contract_list = []
# for index, row in df.iterrows():
# group_list = []
# if row['Unique Key']:
# group_list.append('Unique Key')
# if row['Pricing Before Carveouts']:
# group_list.append('Pricing Before Carveouts')
# if row['Contract Related']:
# group_list.append('Contract Related')
# if row['Provider']:
# group_list.append('Provider')
# if row['Timeline']:
# group_list.append('Timeline')
# if row['Carveout Indicator']:
# group_list.append('Carveout Indicator')
# if row['Carveout Methodology']:
# group_list.append('Carveout Methodology')
# entry_dict = {
# "contract_name": row['Contract Name'],
# "groups": group_list,
# "contract_source_path": "batches/batch_1/"+client+"/"+row['Contract Name']
# }
# contract_list.append(entry_dict)
2024-03-28 17:53:00 +05:30
# contract_list = list(df['Contract Name'])
# myobj = {
# "s3_bucket": 'doczy-dev-infra-textract',
# "batch_id": "1",
# "client_name": client,
# "username": user_mail,
# "contract_list": contract_list
# }
# buttons = st.columns([0.8, 0.2])
# with buttons[0]:
# st.download_button("Download Table", csv, "file.csv", "text/csv", key='download-csv')
# with buttons[1]:
# if st.button("Upload to DB"):
# response = requests.post(doczy_pipeline, json = myobj)
# if response.status_code >= 200 and response.status_code < 300:
# st.write("Success")
# else:
# st.write("Failed")
# # st.write(response.text)